System and method for creating driving route of vehicle

Abstract

A system for creating a driving route of a vehicle includes: a vehicle selecting device to select a surrounding vehicle that is adjacent to a host vehicle; an intention determining device to determine an intention of the surrounding vehicle by using information including a location and a speed of the surrounding vehicle; a driving route predicting device to predict a driving route of the surrounding vehicle based on the determined intention of the surrounding vehicle; a map creating device to create a map by using the predicted driving route of the surrounding vehicle; and a driving route creating device to create a driving route of the host vehicle.

What is claimed is: 
     
       1. A system comprising:
 non-transitory memory storing instructions executable to create a driving route of a vehicle; and 
 a processor configured to execute the instructions to:
 select a surrounding vehicle that is adjacent to a host vehicle; 
 determine an intention of the surrounding vehicle by using information including a location and a speed of the surrounding vehicle; 
 predict a driving route of the surrounding vehicle based on the determined intention of the surrounding vehicle; 
 create a map by using the predicted driving route of the surrounding vehicle; 
 create a driving route of the host vehicle; 
 determine a number of intentions of the surrounding vehicle based on a mean value and a variance, 
 wherein the mean value is an average value of values measured by repeatedly determining the intention of the surrounding vehicle, and the variance is a difference in evaluation values assigned to elements of kinds of the determined intentions of the surrounding vehicle; 
 classify the determined intention of the surrounding vehicle into kinds of a plurality of intentions; 
 calculate evaluation values from the kinds of the plurality of intentions; and 
 select the evaluation values based on difference of variances between the kinds of the plurality of intentions, 
 
 wherein, when determining the intention of the surrounding vehicle, the processor is further configured to:
 collect a set of information regarding the surrounding vehicle including a transverse location of the surrounding vehicle, a longitudinal location of the surrounding vehicle, a transverse speed of the surrounding vehicle, and a longitudinal speed of the surrounding vehicle; and 
 predict the driving route of the surrounding vehicle by feeding a machine learning model with the collected set of information and the selected evaluation values. 
 
 
     
     
       2. The system of  claim 1 , wherein the processor is configured to select the surrounding vehicle connected to the host vehicle by using Delaunay triangulation. 
     
     
       3. The system of  claim 1 , wherein the processor is configured to determine the intention of the surrounding vehicle by using a grid map of surrounding vehicles. 
     
     
       4. The system of  claim 1 , wherein the processor is configured to create the predicted driving route of the surrounding vehicle by generating a circular distribution or an elliptical distribution over time, and configured to create a grid map by using the generated circular or elliptical distribution. 
     
     
       5. The system of  claim 4 , wherein the processor is configured to change a range of the predicted driving route of the surrounding vehicle to be wider as time passes. 
     
     
       6. The system of  claim 1 , wherein the processor is configured to create the driving route of the host vehicle by using a grid map. 
     
     
       7. A computer-implemented method performed by a processor configured to execute instructions to create a driving route of a vehicle stored in non-transitory memory, the method comprising the steps of:
 selecting a surrounding vehicle that is adjacent to a host vehicle; 
 determining an intention of the surrounding vehicle by using information including a location and a speed of the surrounding vehicle; 
 predicting a driving route of the surrounding vehicle based on the determined intention of the surrounding vehicle; 
 creating a grid map by using the predicted driving route of the surrounding vehicle; and 
 creating a driving route of the host vehicle, 
 wherein in determining the intention of the surrounding vehicle includes:
 determining a number of intentions of the surrounding vehicle based on a mean value and a variance, and 
 
 wherein the mean value is an average value of values measured by repeatedly determining the intention of the surrounding vehicle, and the variance is a difference in evaluation values assigned to elements of kinds of the determined intentions of the surrounding vehicle,
 wherein the step of determining the intention of the surrounding vehicle includes: 
 classifying the determined intention of the surrounding vehicle into kinds of a plurality of intentions, 
 calculating evaluation values from the kinds of the plurality of intentions, and 
 selecting the evaluation values based on difference of variances between the kinds of the plurality of intentions, and 
 
 wherein the step of determining the intention of the surrounding vehicle further includes:
 collecting a set of information regarding the surrounding vehicle including a transverse location of the surrounding vehicle, a longitudinal location of the surrounding vehicle, a transverse speed of the surrounding vehicle, and a longitudinal speed of the surrounding vehicle; and 
 predicting the driving route of the surrounding vehicle by providing a machine learning model with the collected set of information and the selected evaluation values. 
 
 
     
     
       8. The method of  claim 7 , wherein the step of selecting the surrounding vehicle includes:
 selecting the surrounding vehicle connected to the host vehicle by using Delaunay triangulation. 
 
     
     
       9. The method of  claim 7 , wherein the step of creating the grid map includes:
 creating the predicted driving route of the surrounding vehicle by generating a circular distribution or an elliptical distribution over time; and 
 creating the grid map by using the generated circular or elliptical distribution. 
 
     
     
       10. The method of  claim 9 , wherein the step of creating the grid map includes:
 changing a range of the predicted driving route of the surrounding vehicle to be wider as time passes.